12.1 Introduction
Head injuries are the most common cause of death and permanent disability in the early decades of life.
They vary widely in their etiology, pathophysiology, clinical presentation, and optimal treatment strate- gies. Traumatic brain injuries are classified in two main categories: focal and diffuse brain injuries [1].
Focal brain injuries usually result from direct impact force to the head, like cerebral contusions and epidur- al hematomas. Diffuse brain injuries are caused by sudden changes in movement of the head, usually ro- tational accelerations, which result in a variety of in- juries, ranging from a brief cerebral concussion to ex- tensive diffuse axonal injuries (DAI).
Computed tomography (CT) is imperative in pa- tients with focal and diffuse injuries, especially when hemodynamically or neurologically unstable [2].
However, CT is often false negative or underestimates contusions shortly after trauma, and DAIs are often not detected. Conventional MR imaging has higher detection sensitivity with regard to these lesions be- cause of its greater sensitivity for edema [3].
Once it was thought that edema following trau- matic brain injury was vasogenic, but recent experi- mental studies using diffusion-weighted (DW) imag- ing have shown that edema after head trauma con- sists of both vasogenic and cytotoxic edema [4–9].
Since DW imaging is also very sensitive in detecting small lesions of cytotoxic edema and can differenti- ate cytotoxic from vasogenic edema, it has become especially useful in the evaluation and staging of pa- tients with DAI.
12.2 Diffuse Axonal Injury
Diffuse axonal injury results from a diffuse shearing- strain deformation causing change in shape of brain tissue from unequal movement of adjacent tissues that differ in density and rigidity [1]. Patients with DAI more often than with other types of primary brain injuries show severe impairment of conscious- ness at impact.
Pathologically, injury related to DAI is always more extensive microscopically than at gross exami- nation [10]. Microscopically, shearing injuries initial- ly produce multiple, characteristic axonal bulbs, or retraction balls, as well as numerous foci of perivas- cular hemorrhages.
The origin behind cytotoxic edema in DAI seems to be related to an excitotoxic mechanism, in partic- ular glutamate [1, 11, 12]. Damage at the node of Ran- vier will result in a traumatic defect in the axonal membrane. This defect causes excessive neurotrans- mitter release with increase in intracellular calcium ions, as in brain ischemia, which leads to axonal and glial cell swelling (cytotoxic or neurotoxic edema).
These changes can eventually lead to axonal degener- ation or necrosis with microglial and astrocytic reac- tive changes. Accumulation of hemosiderin-laden macrophages is also seen in the chronic phase.
12.2.1 Location
Common locations of DAI are at the gray–white mat- ter junctions (Fig. 12.1), in the corpus callosum (Fig.
12.2) and at the dorsolateral aspect of the upper brain stem (Fig. 12.3). DAI may be confined to the white matter of the frontal and temporal lobes in mild head trauma [13]. With more severe rotational accelera- tion, lesions are also seen in the lobar white matter as well as in the posterior half of the corpus callosum. In cases with even greater trauma, lesions will also be found in the anterior corpus callosum, and the dor-
Trauma
solateral aspects of the midbrain and upper pons. Oc- casionally, DAI lesions occur in the parietal and oc- cipital lobes, internal and external capsules (Fig.
12.2), basal ganglia (Fig. 12.4), thalamus (Fig. 12.2), fornix (Fig. 12.5), septum pellucidum, and cerebel- lum (Fig. 12.6). Intraventricular hemorrhage can ac-
company these findings. They have the same me- chanical origin and are due to disruption of the subependymal plexus of capillaries and veins that lie along the ventricular surface of the corpus callosum, fornix, and septum pellucidum [14].
Figure 12.1 a–d
Fig. 12.1a–d.Diffuse axonal injury in gray–white matter junction in a 7-year-old male after a motor vehi- cle accident. T2-weighted (a) and coronal FLAIR (b) images show multiple hyperintense lesions in the gray–white matter junction of bilateral frontoparietal lobes (ar- rows).cCoronal GRE image shows multiple small hemorrhages as low signal in these lesions (ar- rows).dDW image demonstrates diffuse axonal injury as high signal intensity (arrow) with decreased ADC (not shown), representing cy- totoxic edema
Figure 12.3 a–c
Diffuse axonal injury in the brain stem in a 28-year-old male after a motor vehicle accident.aDW image shows a hy- pointense lesion with a hyperintense rim in the dorsolateral aspect of the midbrain, representing a hemorrhagic lesion of diffuse axonal injury (arrow).bADC map shows decreased ADC of this lesion (arrow).This might be due to a paramag- netic susceptibility artifact.cCoronal GRE image clearly shows hemorrhagic lesions as hypointense in the brain stem (ar- row) and in the right frontoparietal region
䉲
Figure 12.2 a–d
Diffuse axonal injury in the corpus callosum, internal capsule and thalamus in a 29-year-old female after a motor vehicle accident. T2- weighted (a) and FLAIR (b) images show multiple hyperintense le- sions in the anterior and posterior corpus callosum, internal capsules and left thalamus (arrows).cDW image demonstrates these lesions as high signal intensity with de- creased ADC (d)
Figure 12.4 a–d
Diffuse axonal injury in the basal ganglia in a 3-year-old male after a motor vehicle accident.a T2- weighted image shows hyperin- tense lesions in the right lentiform and caudate nucleus (arrows).b, c DW imaging shows these lesions as hyperintense with decreased ADC (arrows).dCoronal GRE im- age clearly shows no hemorrhag- ic foci in these lesions
Figure 12.5 a–c
Diffuse axonal injury in the fornix of an 11-year-old female after a motor vehicle accident.aOn T2-weighted image, it is difficult to detect a small hyperintense lesion in the fornix (arrow).b, cDW image shows the lesion in the fornix and pos- terior corpus callosum as hyperintense with decreased ADC (arrows)
䉲
Figure 12.6 a–d
Diffuse axonal injury in the cere- bellum of an 18-year-old male af- ter a motor vehicle accident.aT2- weighted image shows a hypo- intense lesion in the right middle cerebellar peduncle (arrow).bDW image shows a hypointense lesion with a hyperintense rim, repre- senting a hemorrhagic lesion (ar- row).cADC map reveals decreased ADC in this lesion (arrow).This may be due to a paramagnetic suscep- tibility artifact.dCoronal GRE im- age clearly demon- strates hemor- rhagic lesions as hypointense (arrow)
12.2.2 Computed Tomography and MR Imaging
Few DAI lesions are visible with CT. Only large le- sions or those that are grossly hemorrhagic are seen.
MR imaging has been proven to be more sensitive for detection as well as for characterization of DAI le- sions [2]. Conventional MR imaging shows multiple, small, deeply situated elliptical lesions that spare the overlying cortex. Fluid-attenuated inversion-recov- ery (FLAIR) images are more sensitive than T2- weighted images to detect small hyperintense lesions adjacent to the cerebrospinal fluid, such as in the fornix and septum pellucidum [15]. These lesions are, moreover, often accompanied by small, petechial hemorrhages. They occur in 10–30% of all DAI le- sions [16] and are best appreciated on T2*-weighted gradient-echo (GRE) images because of their suscep- tibility effects [17]. However, even these MRI se- quences are thought to underestimate the true extent of DAI.
12.2.3 Diffusion-Weighted Imaging
Diffusion-weighted imaging measures a unique physiologic parameter, movement of water in the tis- sue, which allows for identification of DAI lesions that may not be visible on T2/FLAIR or T2*-weighted GRE images. DAI lesions on DW imaging are hyper- intense and associated with decreased apparent dif- fusion coefficient (ADC) [17–22]. The precise mech- anisms underlying the diffusion changes associated with DAI are unknown. Cytotoxic edema, which seems to be the cause of reduced ADC in ischemic brain injury, can also occur in the early phase of DAI.
However, reduced ADC could also be due to the de-
velopment of retraction balls and concomitant cy- toskeletal collapse along the severed axons [21].
The time course of the ADC abnormality seems to be different from that of ischemic brain injury. Pro- longed decrease in ADC, over 2 weeks, was occasion- ally observed in DAI [18], and cytotoxic edema in the corpus callosum can be partially reversible on fol- low-up imaging using T2-weighted sequences [22].
However, axonal and glial cell swelling in DAI is thought to be mainly due to excitotoxic mechanisms.
It can be a slower or reversible form of cellular swelling than that seen in ischemic brain injuries [6].
Hemorrhagic components, which often accompany these brain injuries, will affect the signal intensity on DW images.
12.3 Brain Contusion
Brain contusions are defined as traumatic injuries to the cortical surface of the brain [1]. They are caused by direct contact between the skull and the brain parenchyma. Compared with DAI, contusions tend to be larger, more superficial, more ill defined and more likely to contain areas of hemorrhage. Cytotoxic ede- ma in brain contusions is also related to excitotoxic mechanisms [23].
12.3.1 Location
Common locations of brain contusions are in the temporal and frontal lobes, especially along their an- terior, lateral, and inferior surfaces (Fig. 12.7). The parietal occipital lobes, hippocampus (Fig. 12.8), cerebellar hemisphere, vermis and cerebellar tonsils (Fig. 12.9) are less frequently involved [2].
Figure 12.8 a–c
Brain contusion in the hippocampus in an 11-year-old female after a motor vehicle accident.aFLAIR image shows a hy- perintense lesion in the left hippocampus (arrow).bDW image shows this lesion as hyperintense.cADC is decreased in the left hippocampus and left side of the brain stem (arrows), representing mainly cytotoxic edema
䉲
Figure 12.7 a–d
Brain contusion in the frontal lobe in a 37-year-old male after a mo- tor vehicle accident.aOn CT ob- tained after evacuation of epidur- al hematoma, it is difficult to delineate the extent of a mass le- sion in the right frontal lobe (ar- rows).bT2-weighted image delin- eates the extent of the edematous brain contusion (arrows).cDW im- age shows heterogeneous signal intensity in these lesions, repre- senting mixed vasogenic and cy- totoxic edema with hemorrhagic necrotic tissues (arrows). d ADC map reveals mixed increase and relative decrease of ADC (arrows) in these lesions
12.3.2 Computed Tomography and MR Imaging
Contusions are often difficult to identify on CT ob- tained shortly after trauma unless they are large or contain areas of hemorrhage [2]. Initial CT will often show only faint areas of low attenuation, sometimes mixed with a few tiny areas of petechial hemorrhage.
MRI is considerably more sensitive than CT for early detection and evaluation of their extent.
12.3.3 Diffusion-Weighted Imaging Findings
Brain contusions are sometimes associated with a non-hemorrhagic mass effect, which progresses rap- idly after the trauma. Edema in brain contusions is heterogeneous, composed of cytotoxic and vasogenic edema [24], which can be demonstrated by DW imag- ing. Kawamata et al. reported a specific DW imaging finding of brain contusions [25, 26]. On DW imaging, the contusion is shown as a low intensity core, with in- creased ADC, surrounded by a rim of a high intensity, with decreased ADC. This suggests that intra- and ex- tracellular components undergo disintegration and homogenization within the central area, whereas cel- lular swelling is predominant in the peripheral area.
12.4 Hemorrhage Related to Trauma
Traumatic hemorrhages result from injury to a cere- bral vessel (artery, vein or capillary) [2]. Subdural hematomas originate from disruption of the bridg- ing cortical veins, which are vulnerable to rapid stretching. Epidural hematoma can have either an ar- terial or a venous sinus origin, typically associ- ated with a skull fracture. Traumatic intracerebral hematomas result from a shear-strain injury involv- ing arteries, veins or capillaries. Traumatic subarach- noid hemorrhage is usually seen after severe head trauma and may as such accompany brain contusion or DAI. Hemorrhages can also represent a disruption of intracranial arteries, especially arteries of the ver- tebrobasilar system.
Figure 12.9 a–c
Brain contusion in the cerebellar tonsil in an 11-year-old female after a motor vehicle accident.aT2-weighted image shows hyperintense lesions in the cerebellar tonsils (arrows).bDW image shows this lesion as hyperintense.cADC is par- tially decreased
12.4.1 Computed Tomography and MR Imaging
Computed tomography is the modality of choice for the initial evaluation of traumatic brain hemorrhag- es, as it is a fast examination technique, is widely available, has no contraindications and relatively ac- curately depicts most hematomas [2].
Magnetic resonance imaging is usually not the primary imaging technique and findings will then depend on the stage of degradation of the hemoglo- bin at the time of examination. However, in most in- stances MR imaging is extremely helpful to detect hematomas, especially along the vertex and skull base, and can in certain questionable cases differenti- ate between subdural and epidural hematomas [2].
T2*-weighted GRE and FLAIR images seem to be more sensitive to detect hemorrhage than conven- tional spin-echo imaging [27–30].
12.4.2 Diffusion-Weighted Imaging
Diffusion-weighted imaging findings of subdural and epidural hematomas have not been well described in the literature. Depending on the age of the hematoma, DW imaging will vary in signal intensity (Fig. 12.10).
Gradient-echo sequences are better in detecting hematomas, including subdural and epidural hematomas, than DW imaging [30]. Although often difficult to detect [31, 32], DW imaging can occasion- ally depict a subarachnoid hemorrhage as a hyperin- tense signal (Fig. 12.11). The benefit of DW imaging is probably for the detection of underlying or associated parenchymal lesions. For example, subarachnoid hemorrhage will often cause vasospasm of the in- tracranial arteries, which can result in brain ischemia.
Mass effect secondary to subdural or epidural hematomas, which is closely related to morbidity and mortality, is due to a combination of the hematoma, underlying parenchymal edema and diffuse cerebral swelling.
Figure 12.10 a–d
Epidural and subdural hematoma in a 26-year-old male after a mo- tor vehicle accident.aCT shows a left epidural hematoma (arrow) but it is difficult to depict the iso- dense small subdural hematoma in the right side (arrowheads).
bT2-weighted image shows the left epidural hematoma (arrow) as a hypointense lesion and the right subdural hematoma as partially hypointense lesions (arrowheads).
cDW image shows the epidural hematoma as very hypointense due to deoxy-hemoglobin, and the subdural hematoma as very hyperintense presumably due to high viscosity or hypercellularity of hematoma.dADC map shows hypointensity due to loss of pixels with background masking in the left epidural hematoma (arrow).
ADC map also shows decreased ADC in the right subdural hema- toma (arrowheads)
Figure 10.11 a–d
Subarachnoid hemorrhage in a 68-year-old male with ruptured aneurysm of the right middle cerebral artery bifurcation.aPost- operative CT shows subtle high density of subarachnoid space in the right frontoparietal area (ar- rows).bFLAIR image shows sub- arachnoid hemorrhage as hyper- intensity.cDW image also shows subarachnoid hemorrhage as hy- perintensity with mildly increased ADC (not shown).dCoronal GRE shows the hemorrhage as low sig- nal intensity
12.5 Vascular Injuries
Traumatic arterial and venous injuries (dissections, lacerations, occlusions, pseudoaneurysm, arteriove- nous fistulas) are more prevalent than generally be- lieved [2]. Many asymptomatic lesions probably es- cape detection, and others are recognized several days to months after the injury (Fig. 12.12). CT is use- ful to detect skull base fractures and CT angiography
may help to evaluate the vascular injuries. However, a combination of MR imaging and MR angiography is probably the most efficacious way to screen high-risk patients for traumatic vascular injuries, especially if combined with DW imaging, which is very sensitive to detect small and early ischemic lesions secondary to traumatic vascular injuries. Still, one has to ac- knowledge that conventional angiography continues to be the gold standard in the evaluation of known or suspected traumatic arterial lesions.
Figure 12.12 a–c
Cerebral infarction after carotid artery dissection with pseudoaneurysm in a 20-year-old female after a motor vehicle ac- cident.aT2-weighted image shows hyperintense lesions in the right middle cerebral artery territory including the right basal ganglia (arrows).bDW image also shows these lesions as hyperintense with decreased ADC (not shown), repre- senting acute infarction.cConventional angiogram shows pseudoaneurysm of the right carotid artery (arrow)
References
1. Gennarelli TA (1993) Mechanisms of brain injury. J Emerg Med 11 Suppl 1:5–11
2. Gentry LR (1994) Imaging of closed head injury. Radiolo- gy 191:1–17
3. Kelly AB, Zimmerman RD, Snow RB, Gandy SE, Heier LA, Deck MD (1988) Head trauma: comparison of MR and CT- experience in 100 patients. AJNR Am J Neuroradiol 9:699–708
4. Alsop DC, Murai H, Detre JA, McIntosh TK, Smith DH (1996) Detection of acute pathologic changes following ex- perimental traumatic brain injury using diffusion-weighted magnetic resonance imaging. J Neurotrauma 13:515–521 5. Hanstock CC, Faden AI, Bendall MR, Vink R (1994) Diffu-
sion-weighted characterization, edema, diffusion aging differentiates ischemic tissue from traumatized tissue.
Stroke 25:843–848
6. Barzo P, Marmarou A, Fatouros P, Hayasaki K, Corwin F (1997) Contribution of vasogenic and cellular edema to traumatic brain swelling measured by diffusion-weighted imaging. J Neurosurg 87:900–907
7. Ito J, Marmarou A, Barzo P, Fatouros P, Corwin F (1996) Characterization of edema by diffusion-weighted imaging in experimental traumatic brain injury. J Neurosurg 84:97–103
8. Albensi BC, Knoblach SM, Chew BG, O’Reilly MP, Faden AI, Pekar JJ (2000) Diffusion and high resolution MRI of trau- matic brain injury in rats: time course and correlation with histology. Exp Neurol 162:61–72
9. Assaf Y, Beit-Yannai E, Shohami E, Berman E, Cohen Y (1997) Diffusion- and T2-weighted MRI of closed-head in- jury in rats: a time course study and correlation with his- tology. Magn Reson Imaging 15:77–85
10. Gennarelli TA, Graham DI (1998) Neuropathology of the head injuries. Semin Clin Neuropsychiatry 3:160–175 11. Faden AI, Demediuk P, Panter SS,Vink R (1989) The role of
excitatory amino acids and NMDA receptors in traumatic brain injury. Science 244:798–800
12. Katayama Y, Becker DP, Tamura T, Ikezaki K (1990) Cellu- lar swelling during cerebral ischaemia demonstrated by microdialysis in vivo: preliminary data indicating the role of excitatory amino acids. Acta Neurochir Suppl (Wien) 51:183–185
13. Mittl RL, Grossman RI, Hiehle JF, Hurst RW, Kauder DR, Gennarelli TA, Alburger GW (1994) Prevalence of MR evi- dence of diffuse axonal injury in patients with mild head injury and normal head CT findings. AJNR Am J Neurora- diol 15:1583–1589
14. Gentry LR, Thompson B, Godersky JC (1988) Trauma to the corpus callosum: MR features. Am J Neuroradiol 9:1129– 1138
15. Ashikaga R, Araki Y, Ishida O (1997) MRI of head injury using FLAIR. Neuroradiology 39:239–242
16. Yanagawa Y, Tsushima Y, Tokumaru A, Un-no Y, Sakamoto T, Okada Y, Nawashiro H, Shima K (2000) A quantitative analysis of head injury using T2*-weighted gradient-echo imaging. J Trauma 49:272–277
17. Huisman TA, Sorensen AG, Hergan K, Gonzalez RG, Schae- fer PW (2003) Diffusion-weighted imaging for the evalua- tion of diffuse axonal injury in closed head injury. J Com- put Assist Tomo 27:5–11
18. Liu AY, Maldjian JA, Bagley LJ, Sinson GP, et al. (1999) Trau- matic brain injury: diffusion-weighted MR imaging find- ings. AJNR Am J Neuroradiol 20:1636–1641
19. Rugg-Gunn FJ, Symms MR, Barker GJ, Greenwood R, Dun- can JS (2001) Diffusion imaging shows abnormalities after blunt head trauma when conventional magnetic resonance imaging is normal. J Neurol Neurosurg Psychiatry 70:530–533
20. Hergan K, Schaefer PW, Sorensen AG, Gonzalez RG, Huis- man TA (2002) Diffusion-weighted MRI in diffuse axonal injury of the brain. Eur Radiol 12:2536–2541
21. Arfanakis K, Haughton VM, Carew JD, Rogers BP, Dempsey RJ, Meyerand ME (2002) Diffusion tensor MR imaging in diffuse axonal injury. AJNR Am J Neuroradiol 23:794–802 22. Takayama H, Kobayashi M, Sugishita M, Mihara B (2000)
Diffusion-weighted imaging demonstrates transient cyto- toxic edema involving the corpus callosum in a patient with diffuse brain injury. Clin Neurol Neurosurg 102:135–139
23. Maeda T, Katayama Y, Kawamata T, Yamamoto T (1998) Mechanisms of excitatory amino acid release in contused brain tissue: effects of hypothermia and in situ adminis- tration of CO2+ on extracellular levels of glutamate. J Neu- rotrauma 15:655–664
24. Umeda M, Yamaki T, Tanaka C, et al. (1996) Analysis of brain edema in experimental cerebral contusion using dif- fusion weighted MRI and ADC value. Neurotraumatology 19:79–83
25. Kawamata T, Katayama Y, Mori T, Aoyama N, Tsubokawa T (2002) Mechanisms of the mass effect of cerebral contu- sion: ICP monitoring and diffusion MRI study. Acta Neu- rochir Suppl 81:281–283
26. Kawamata T, Katayama Y, Aoyama N, Mori T (2000) Het- erogeneous mechanisms of early edema formation in cere- bral contusion: diffusion MRI and ADC mapping study.
Acta Neurochir Suppl 76:9–12
27. Melhem ER, Jara H, Eustace S (1997) Fluid-attenuated in- version recovery MR imaging: identification of protein concentration thresholds for CSF hyperintensity. Am J Roentgenol 169:859–862
28. Singer MB,Atlas SW, Drayer BP (1998) Subarachnoid space disease: diagnosis with fluid-attenuated inversion-recov- ery MR imaging and comparison with gadolinium-en- hanced spin-echo MR imaging–blinded reader study. Ra- diology 208:417–422
29. Dechambre SD, Duprez T, Grandin CB, Lecouvet FE, Peeters A, Cosnard G (2000) High signal in cerebrospinal fluid mimicking subarachnoid haemorrhage on FLAIR following acute stroke and intravenous contrast medium.
Neuroradiology 42:608–611
30. Taoka T, Yuh WT, White ML, Quets JP, Maley JE, Ueda T (2001) Sulcal hyperintensity on fluid-attenuated inversion recovery MR images in patients without apparent cere- brospinal fluid abnormality. AJR Am J Roentgenol 176:
519–524
31. Lin DD, Filippi CG, Steever AB, Zimmerman RD (2001) De- tection of intracranial hemorrhage: comparison between gradient-echo images and b0 images obtained from diffu- sion-weighted echo-planar sequences. AJNR Am J Neuro- radiol 22:1275–1281
32. Wiesmann M, Mayer TE, Yousry I, Medele R, Hamann GF, Bruckmann H (2002) Detection of hyperacute subarach- noid hemorrhage of the brain by using magnetic reso- nance imaging. J Neurosurg 96:684–689
